A Multifactorial Risk Scoring System for the Prediction of Early Relapse in CMML Patients with Allo-Hsct: a Nationwide Representative Multicenter Study

Jian-Ying Zhou,Yu-Xiu Chen,Hai-Long Yuan,Ya-Jing Xu,Xiao-Bing Huang,Su-Jun Gao,Yi-Cheng Zhang,Fang Zhou,Xian-Min Song,Yi Luo,Jian-Min Yang,Yu-Hua Li,Shun-Qing Wang,Yu-Jun Dong,Xi Zhang,Yi-Mei Feng,Xin Du,Han Zhu,Zun-Min Zhu,Ke-Hong Bi,Ming Jiang,Ting Niu,Ding-Ming Wan,Yi Chen,Li Liu,Hai Yi,Yu-Hong Chen,Feng-Rong Wang,Yuan-Yuan Zhang,Xiao-Dong Mo,Wei Han,Jing-Zhi Wang,Yu Wang,Huan Chen,Xiang-Yu Zhao,Ying-Jun Chang,Kai-Yan Liu,Xiao-Jun Huang,Xiao-Hui Zhang
DOI: https://doi.org/10.1038/s41409-024-02480-3
2024-01-01
Bone Marrow Transplantation
Abstract:Chronic myelomonocytic leukemia (CMML) is a clonal hematopoietic stem cell malignancy and the only curable therapy is allogeneic hematopoietic stem cell transplantation (allo-HSCT). However, allo-HSCT is not appropriate for all CMML patients, and relapse is the leading cause of treatment failure. This project conducted a nationwide multicenter real-world study to develop a novel prediction scoring system for early relapse. A total of 238 CMML patients from twenty-seven medical centers treated with allo-HSCT, and 307 adult patients with CMML who underwent allo-HSCT in a publicly available research dataset from the Center for International Blood and Marrow Transplantation Registry (CIBMTR) database were included. Independent prognostic factors for the early relapse of CMML posttransplantation were identified according to competing risk regression methods. Four prognostic factors were identified: bone marrow blasts >10% (hazard ratio [HR], 4.262; P = 0.014), age >60 years (HR, 6.221; P = 0.007), hemoglobin level <100 g/L (HR, 3.695; P = 0.004), and non TET2 gene mutation (HR, 3.425; P = 0.017). A risk-grading scoring system was developed based on the regression coefficients and patients were stratified into low-risk (0-1 point), intermediate-risk (1.5-2 points) and high-risk ( > 2 points) groups. The validated internal c-statistic was 0.767 (95% confidence interval [CI], 0.674-0.860), and the external c-statistic was 0.769 (95% CI, 0.703-0.836). In the derivation cohort, the cumulative incidence rates of early relapse in the low-risk, intermediate-risk, and high-risk groups were 1.35% (95% CI: 1-4%), 10.40% (95% CI: 4-16%), and 29.54% (95% CI: 16-39%) (P < 0.001), respectively. This scoring system can be utilized to early identification of patients at a high risk of relapse and contributing to the implementation of urgent medical support.
What problem does this paper attempt to address?